Neuro Language Generator
Journal Title: International Journal on Computer Science and Engineering - Year 2010, Vol 2, Issue 5
Abstract
‘Neuro Language Generator using Finite State Machine’ is based on neural network and finite state machine. The undamental properties of neural network along with the power of Turing machine prove how it can be implemented for formal language processing. This paper laborates the conventional dynamical language generators, limitations of the conventional dynamical language generators and proposes a new rchitecture for formal language processing. The conventional dynamical language generators used for neural language generators is feedforward RNN. It expresses dynamical language generator using finite automaton and dynamical language enerator using pushdown automaton. Conventional dynamical generators tend to have stability problem, incapable of network training and lack of memory. It is roposed that the new method ‘Neuro Language Generator using Finite State Machine’ solves most of the problems, which the traditional ethods fail to do. The approach employs finite state technology for a RNN in the task of learning to achieve stability in network structure. RNN architecture performs the same computation as a Turing machine. The RNN architecture acts as a language generator, which accepts formal language. Neuro Language Generator is a RNN that uses feedback connections. NLG can be used to solve more omplicated problems compared to traditional dynamical generator.
Authors and Affiliations
P. Dinadayalan , Gnanambigai Dinadayalan , R. Vasantha Kumari
Keywords
Related Articles
- EP ID EP155308
- DOI -
- Views 123
- Downloads 0
Applying moving boundary to compute the alluvium transport and testing the results by remote sensing
To compute the alluvium transport, we use a current model based on a 2D finite-difference grid and a sediment transport model. The first one gives the velocity distribution on the surface of water body and in the case of...
EFFICIENT TRANSACTION REDUCTION IN ACTIONABLE PATTERN MINING FOR HIGH VOLUMINOUS DATASETS BASED ON BITMAP AND CLASS LABELS
Frequent pattern mining in databases plays an indispensable role in many data mining tasks namely, classification, clustering, and association rules analysis. When a large number of item sets are processed by the databas...
Self-Healing in Dynamic Web Service Composition
Web service composition is defined as an orchestration of multiple web services into a single composite web service. Web service composition is done in three ways such as static web service composition, dynamic web servi...
Statistical Approach to Transliteration from English to Punjabi
Machine transliteration plays an important role in natural language applications such as information retrieval and machine translation, especially for handling proper nouns and technical terms. Transliteration is a cruci...
Combining Speedup Techniques based on Landmarks and Containers
The Dijkstra’s algorithm [1] , which is applied in many real world problems like mobile routing, road maps, railway networks, etc,. is used to find the shortest path between source and destination. There are many techniq...